Front page | perl.perl5.porters |
Postings from January 2013
This post is not to discredit anyones RE work. I recently saw some discussion somewhere about "bus factor" and Perl. So I've been thinking about where Perl 5 is going, and specifically the RE engine. I predict this email/thread will be mostly philosophy and opinions without code, so if you have real work to do, go back to it. When Perl got REs (1980s/1990s), I dont think there were any FOSS RE libraries to copy/embed. Now in the 2010s, everyone has an RE engine ( http://en.wikipedia.org/wiki/Comparison_of_regular_expression_engines ). So here are some questions for thought I'd like to see answers/opinions to. I don't have an opinion on any of them. Is Perl 5 maintaining its own RE a good use of limited developer resources? are there limited developer resources in the first place? When P5 RE implements new syntax, does other RE engines still follow? Does P5 RE follow other's RE new syntax? What is the speed (I won't define that) of P5 RE vs other RE in other languages? What if someone skilled in the RE engine from p5p is hit by a bus? who is left? are there enough docs and code comments for someone else to take over? Would you replace Perl's native RE engine, and if so, with what other engine? What would be the challenges of replacing the native Perl RE engine? Should the P5 RE engine spun off as a dual life?Thread Next